A client request control method and device

Abstract

The present invention discloses a client request control method and device, wherein the method includes: obtaining client historical request amount change curve by monitoring client request; obtaining each time period according to the client historical request amount change curve The change rate of the request volume above is used to obtain the next request time interval of the client, and the request control is performed on the requesting client through the next request time interval. The invention can not only ensure that the server normally returns the data requested by the client, but also save resources of the server and enable the client to obtain higher data update frequency.

Description

technical field [0001] The present invention relates to Internet technology, in particular to a client request control method and device. Background technique [0002] At present, the prior art generally adopts a timing request scheme or a simple feedback scheme. Among them, the so-called timing request scheme means that the client ignores the status of the server, and initiates a request to the server every fixed time period indicated by the server or at each fixed time point indicated by the server, and the server does not feedback the status of the server to the client. Such as the current load situation. The so-called simple feedback scheme means that the server evaluates how to provide services to the client according to the status of the server, such as the current load situation. If the data is allowed to be returned through calculation and evaluation, the server returns the data required by the client, or if it is evaluated through calculation. If data is not allow...

AI Technical Summary

Problems solved by technology

[0004] 1) The timing request scheme is adopted, because the client usually sends a request to the server according to the fixed time interval indicated by the server or each fixed time point indicated by the server, and the fixed time interval indicated by the server or each fixed time point does not follow the server Therefore, during the peak period of client requests, a large number of clients send requests to the server, and the server may receive requests exceeding the maximum safe load, which will cause an avalanche effect and lead to a decrease in system throughput , so that the server cannot normally return the data requested by the client; and during the trough period of client requests, only a small number of clients send requests to the server, and a large number of resources on the server are idle, but the client cannot send requests to the server at a higher frequency. The server sends requests to increase the data update frequency, but the bandwidth and server utilization are not sufficient, which not only wastes server resources, but also the client cannot obtain higher data update frequency

[0005] 2) The simple feedback scheme is adopted, because the server evaluates how to provide services for the client according to the state of the server, such as the current load situation, and may or may not return data to the client. Therefore, the simple feedback method When the server load is high and client requests peak, it is easy to cause some clients to return without success many times; while in the client request trough period, the utilization of bandwidth and server is also very insufficient, wasting server resources

Images